iFi Audio Nano iDSD discussion + impression
Jun 24, 2014 at 1:42 PM Post #648 of 2,063
  I contacted iFi and Avatar Acoustics, still no response how to get both...

 
Hi,
 
Best you pm us as we clear our emails on a daily basis!
 
So your email may not have made it through to us.
 
Cheers.
 
iFi audio Stay updated on iFi audio at their sponsor profile on Head-Fi.
 
https://www.facebook.com/people/IFi-audio/61558986775162/ https://twitter.com/ifiaudio https://www.instagram.com/ifiaudio/ https://ifi-audio.com/ https://www.youtube.com/@iFiaudiochannel comms@ifi-audio.com
Jun 24, 2014 at 1:45 PM Post #649 of 2,063
Hi, I have trouble in configuring ifi nano iDSD in foobar2000 so that also in ABX comparator it would play natively DSD.
 
When I play the files normally, it will do DSD256, under ABX comparator it switches DSD playback to 176/192/24 - making direct blind testing of say redbook CD WAV 44.1/16 to DSD impossible. I tried as I might to listen blind. and after I got consistently WRONG results, did allow to see which colour LED is glowing - and discovered the above.
 
Any ideas ?
 
Jun 25, 2014 at 5:50 PM Post #650 of 2,063
I haven't tried iDSD with latest firmware yet (should I try? ...or be afraid of doing so...) Anyways, I know that iDSD (at least mine) does not switch smooth between DSD and PCM under Foobar2k with foo_output_dsd. I don't know exact scenario, but I know that when I double click on PCM record and then on DSD and the on PCM, iDSD sometimes goes crazy, and I have to switch it off and unplug USB cable.

So, from this experience I think ABX comparator may not work good with iDSD, as it probably wants to switch between PCM and DSD at random times.
 
Jun 26, 2014 at 1:42 AM Post #651 of 2,063
You should try the nano with the latest firmware - it includes playback of DSD256 files. Something totally unheard of in this price range IIRC. The true DSD256 (and not upconverted ) recordings are still extremely rare and including of DSD256 playback capability is like something of putting the cart before the horse - as of present, the only commercially available DSD256 recorder is
Merging Horus http://www.merging.com/products/horus Clicking on this link made my day - as I just discovered
the little son of Horus, the Hapi http://www.merging.com/products/hapi This appears to be a trickle-down  "improved" model of (hopefully) lower price, the prime advantage over Horus is its capability to be powered from all kinds of power supplies, including the use of pure DC supply - the battery 
L3000.gif
.
 
You are experiencing similar troubles I had with the nano - prior to "un-hogging" the computor ( registry clean up, some DLL repair, etc, etc, something my computer guru had to take care of ). The nano now works with regular DSD and PCM playback and switching from one to another and vice versa just fine, with the exception this transition is accompanied by some low volume "clicks and noise" that last around one second overall; they are not silent and at least give away there is switching from DSD to PCM or vice versa going on.
 
Under ABX comparator, nano can get crazy. Besides losing native DSD capability, which is more likely a foobar2000 limitation/problem,  switching from playing back a DSD and DXD file can result in music being accompanied by horrific mainly high frequency noise - VERY loud, close to 0 dBFS. It was not a pleasent surprise when ABing the same track in DSD and DXD using IEMs ...
I expect this to be a problem of the limited capabilities of my PC - only 4GB of RAM and processor at 2.33 MHz.
Stopping the playback restores the normal operation in my case, no disconnection of the USB connection required. 
As always with computers, YMMV.
 
Jun 26, 2014 at 12:08 PM Post #652 of 2,063
I have tried some DSD content via some free sample downloads etc. I'm not a classical fan so the whole DSD thing is kind of mute.
 
Its hard to evaluate how 'good ' the DSD sounds for me because I cant evaluate the music that I'm not familiar with. It sounds 'good' to me but not better than a good HDCD of music that I like, that was ripped to 24 bit using dbpoweramp etc. Again in terms of content its apples and oranges.
 
I havent bought any of the rock DSD content available yet (ripped from SACD). That would be a better test, but I cant justify spending $30 a pop for music I already have, although buying the Zep remasters in 24/96 is tempting. I think the jury is still out on those...
 
Jun 26, 2014 at 4:08 PM Post #653 of 2,063
I have noticed the biggest difference in the way violin sounds. In the PCM mode they sounded constrained by some plastic cage, while in DSD mode they sounded unconstrained and very clean. It didn't really matter if source material was DSD or PCM converted on the fly to DSD, I heard no difference what so ever. I also couldn't hear any difference between 96kHz and 352kHz PCM - they both sounded to me the same. The difference was only audible between iDSD being physically in PCM mode or DSD mode. In this tests I used Alpha Dogs powered by Pan Am. These two actually couple very nice with iDSD.
 
Jun 26, 2014 at 6:35 PM Post #654 of 2,063
I have noticed the biggest difference in the way violin sounds. In the PCM mode they sounded constrained by some plastic cage, while in DSD mode they sounded unconstrained and very clean. It didn't really matter if source material was DSD or PCM converted on the fly to DSD, I heard no difference what so ever. I also couldn't hear any difference between 96kHz and 352kHz PCM - they both sounded to me the same. The difference was only audible between iDSD being physically in PCM mode or DSD mode. In this tests I used Alpha Dogs powered by Pan Am. These two actually couple very nice with iDSD.

 
How would you compare the onboard DAC of the Pan Am to the iDSD Nano's
 
Jun 27, 2014 at 4:35 AM Post #655 of 2,063
I would say iDSD sounds significantly cleaner, more precise, more detailed. But still if I were to arrange how I rate sound quality between devices I have (or had) this is the way I would do it:

Xperia X8 mini < iPhone 4 < Moto G < K850i ~ N900 < Lenovo Y580 < Xonar DG ~ Pan Am < iDSD < NAD T744

I asked NAD guys about the DAC, and they said it's Cirrus Logic 96kHz/24bit, though I don't know which exact chip (I would have to open up the chasis). I can guess it probably is a codec like this one: http://www.cirrus.com/en/products/cs4244.html. Anyways I rate the SQ as more human-ear-friendly than iDSD. The iDSD sounds better (to me) only when in true DSD mode.
 
Jun 27, 2014 at 12:07 PM Post #656 of 2,063
Dont mean to be a fly in the ointment to the resident experts on this site, but I have, yet again, confirmed the superior sound of redbook upsampled to 2xDSD vs playing native. I can qualify the difference, although very subtle yet very noticeable with very good recordings (Dire Straits) when I sit right in front of the speaker. Native 16/44.1 has a bit of a digital nature that is somewhat irritating, and less real (sort of a digital hash). There is a 'smoothness' to the upsampled sound. As others have said this may have to do with the way the iDSD Nano processes DSD vs. PCM. I also tried upsampling the PCM to 384khz with no real difference from the native 44.1.
 
Looking forward to the reviews on the new Micro when it comes out!
 
Jun 27, 2014 at 6:18 PM Post #657 of 2,063
Dont mean to be a fly in the ointment to the resident experts on this site, but I have, yet again, confirmed the superior sound of redbook upsampled to 2xDSD vs playing native. I can qualify the difference, although very subtle yet very noticeable with very good recordings (Dire Straits) when I sit right in front of the speaker. Native 16/44.1 has a bit of a digital nature that is somewhat irritating, and less real (sort of a digital hash). There is a 'smoothness' to the upsampled sound. As others have said this may have to do with the way the iDSD Nano processes DSD vs. PCM. I also tried upsampling the PCM to 384khz with no real difference from the native 44.1.

Looking forward to the reviews on the new Micro when it comes out!


This is exactly what I experience, just didn't know how to name it :D I'm glad I'm not the only one who feels iDSD plays best DSD :)
 
Jun 28, 2014 at 6:31 AM Post #658 of 2,063
So, I know I'm going to get a lot of flak for this...Yes, I'm new here (to posting at least, but not really to reading) - but really, I hope I'm soundly somehow proven wrong.
 
I just got my iFi Audio Nano iDSD, and I've had like.. less than a day to play with it so far, but something is bothering me.
 
This device claims to be able to do DSD128 and PCM 32/384 and DXD.  And with the newer firmware, DSD256.
That's all really really awesome... if that's what it could do.
 
See, it's become clear that the iFi Audio Nano iDSD uses a DSD1793 DAC chip. While that's not a horrible chip by any means, it sure as heck can't do what iFi/AMR is claming it can do. So what's going on here?
 
The DSD1793 can do DSD64, and PCM 24/192, and sorry -- that's all she wrote.  (Okay, okay; PCM 24/200, but who's counting.)
 
This is in front of everyone here. Disassembly shots of the iDSD Nano are available, and the datasheet for the DSD1793 is available.
So are we all operating on some level of collective cognitive dissonance, or am I missing something here?
 
Is iFi/AMR somehow "overclocking the dac" (and "overwording it")?  Unlikely. Really unlikely.
I'd say more likely, that this dac is outputting DSD64/2.8mhz and PCM 24/192, no matter what it's being fed. 
 
I'm guessing with these higher signal rates, a LOT of data is either being thrown away in the driver, or by something going on in the XMOS USB chip.
I'm really surprised this hasn't been talked about or addressed.
 
And yeah, I expect flames after this, but, "so be it" - the DSD1793 can *not do* DSD128, DSD256, DXD, or PCM 32-bit (of any sort) or PCM 352 or 384 khz (of any sort.)
So, i'ts *got to* be truncating or downconverting.
 
I'm not saying it's doing a bad job at it, but this is starting to remind me of the old modem days when "57600 BPS!" was advertised; when that was just the data input rate, (or serial port rate), but the actual connection rate was much, much slower.
 
I hope this isn't the trend we're going down with DACs. But, for $189, I figured, "DSD128 and DSD256?? For that price?? Too good to be true."
 
Unfortunately, it looks like it is. I'm *NOT* here to troll.  If you guys can ring my bell on this and set me straight, by all means, do so. I'll happily and enthusiastically be proven wrong. I really, really, REALLY want to like this DAC.
 
But if AMR is telling tall tales, it's gonna leave a bad taste in my mouth...
 
Jun 28, 2014 at 6:47 AM Post #659 of 2,063
  See, it's become clear that the iFi Audio Nano iDSD uses a DSD1793 DAC chip. While that's not a horrible chip by any means, it sure as heck can't do what iFi/AMR is claming it can do. So what's going on here?
 
... But if AMR is telling tall tales, it's gonna leave a bad taste in my mouth...

 
I'll leave the detail to iFi Audio to answer, if they are willing to go into the detail that is, since these are more or less their trade secret for now and they probably don't want others to copy what they have done. Hence why they are being a bit cryptic on this discussion.
 
I asked them the same question awhile back, and the answer is yes, the DSD1793 can't do DSD128, DSD256, DXD, etc if you use the default configuration that TI tells you to use. But deep inside the silicone, the chip itself is actually capable of native decoding for those formats. The way iFi is able to explore the DSD1793's full potential is by employing a customized XMOS implementation that allows the DAC to do those native decoding function that are otherwise hidden by TI.
 
Jun 28, 2014 at 6:50 AM Post #660 of 2,063
  So, I know I'm going to get a lot of flak for this...Yes, I'm new here (to posting at least, but not really to reading) - but really, I hope I'm soundly somehow proven wrong.
 
I just got my iFi Audio Nano iDSD, and I've had like.. less than a day to play with it so far, but something is bothering me.
 
This device claims to be able to do DSD128 and PCM 32/384 and DXD.  And with the newer firmware, DSD256.
That's all really really awesome... if that's what it could do.
 
See, it's become clear that the iFi Audio Nano iDSD uses a DSD1793 DAC chip. While that's not a horrible chip by any means, it sure as heck can't do what iFi/AMR is claming it can do. So what's going on here?
 
The DSD1793 can do DSD64, and PCM 24/192, and sorry -- that's all she wrote.  (Okay, okay; PCM 24/200, but who's counting.)
 
This is in front of everyone here. Disassembly shots of the iDSD Nano are available, and the datasheet for the DSD1793 is available.
So are we all operating on some level of collective cognitive dissonance, or am I missing something here?
 
Is iFi/AMR somehow "overclocking the dac" (and "overwording it")?  Unlikely. Really unlikely.
I'd say more likely, that this dac is outputting DSD64/2.8mhz and PCM 24/192, no matter what it's being fed. 
 
I'm guessing with these higher signal rates, a LOT of data is either being thrown away in the driver, or by something going on in the XMOS USB chip.
I'm really surprised this hasn't been talked about or addressed.
 
And yeah, I expect flames after this, but, "so be it" - the DSD1793 can *not do* DSD128, DSD256, DXD, or PCM 32-bit (of any sort) or PCM 352 or 384 khz (of any sort.)
So, i'ts *got to* be truncating or downconverting.
 
I'm not saying it's doing a bad job at it, but this is starting to remind me of the old modem days when "57600 BPS!" was advertised; when that was just the data input rate, (or serial port rate), but the actual connection rate was much, much slower.
 
I hope this isn't the trend we're going down with DACs. But, for $189, I figured, "DSD128 and DSD256?? For that price?? Too good to be true."
 
Unfortunately, it looks like it is. I'm *NOT* here to troll.  If you guys can ring my bell on this and set me straight, by all means, do so. I'll happily and enthusiastically be proven wrong. I really, really, REALLY want to like this DAC.
 
But if AMR is telling tall tales, it's gonna leave a bad taste in my mouth...

Ifi has obviously tested the DSD1793, along probably pretty much any other chip, for what it can do on paper - and beyond. They must have found out some way to squeeze out the DSD256 out of DSD1793; they did say they used some features not mentioned by Burr Brown/Ti in their data sheet.
 
Now it would be interesting to see to how much of this will be made public - provided it really can do DSD256 in the first place. Foobar2000 with nano clearly recognizes DSD256 as such. JRiver 19 does not indicate what is it playing directly; only kbps is displayed. For DSD128 it is 11289 kbps, for DSD256 it is 22579 kbps - which is exactly the same kbps for the DXD352.8 kHz .
 

Users who are viewing this thread

Back
Top